The proposed work addresses the following issue: although existing biological databases are individually well-curated and interconnected through links to records in other databases, navigation between those records remains a time-intensive, manual process. A biologist may want to know, for example, what geriatric-related proteins have been identified and what knowledge about those proteins has been collected by other researchers over, say, the past two years. To answer this, the biologist must identify relevant data sources and then manually query and integrate answers from each individual source. Although the query process can be automated with scripts, extending or altering the process later requires writing new scripts. Further, scripts are limited in the degree to which they capture biological expertise so that it can be reused for future related queries. Finally, unless explicitly written to do so, scripts do not assist the user in filtering retrieved data and resolving inconsistencies. Our proposed work seeks to develop computational infrastructure that uses XML-based web technology and database/mediation technology. This infrastructure will enable a user to answer questions that require data from multiple sources without explicit user knowledge of source-specific details including location, data structure, and computational interface. The same infrastructure that enables the user to pose and execute such queries will also facilitate users to apply their expertise to filter, rank, and resolve inconsistencies in the query results. The outcome of the research will be designing the infrastructure to any computational discovery task that requires access to multiple, heterogeneous data sources. We achieve these objectives through the following resarch aims: Aim 1 : Design a mediation infrastructure for research on the Genetics and Proteomics of aging. Aim 2: Enhance the access to information related to age-related diseases, disorders and disabilities through the exploration of multiple data sources.